This code is for the caregiver speech measures from transcripts used in the manuscript.

Load libraries and set theme

library(tidyverse)
library(Hmisc)
library(GGally)
library(ppcor)
library(gridExtra)
library(psych)

# https://github.com/ggobi/ggally/issues/139
my_custom_smooth <- function(data, mapping, ...) {
  ggplot(data = data, mapping = mapping) +
    geom_point(alpha = .4, color = I("black")) + 
    geom_smooth(method = "lm", color = I("blue"), ...)
}

theme_set(theme_bw())

Read in data

# NOTE about periods of non-tCDCS
# gemods refers to when there are designated start/end periods of other-directed speech (ODS); this was captured using gems (@G) using CHAT conventions
# kwalods refers to when ODS was transcribed at an utterance-level within a tCDS activity period between caregiver and child (e.g., other-directed speech in the background); this was captured per utterances using CHAT postcodes
## for tokens/min and types/min, we do not include ODS that occurred within a period of tCDS, because durations were captured by activity and not by utterance
## for mlu, we include all ODS across gemods and kwalods


# NOTE about speech == "all"
# "speech" includes two levels: all, spont
# all = refers to all speech by caregivers
# spont = refers to only speech by caregivers that was considered spontaneous rather than recited (e.g., reading book text, singing memorized common songs like itsy bitsy spider); therefore, 'spont' is a subset of 'all'

# freq
freq <- read_csv("./data_demo_lena_transcripts/freq.csv") %>% 
  dplyr::select(-"...1") %>% 
  filter(activity != "kwalods") %>% 
  filter(speech == "all") %>% 
  mutate(activity = factor(activity, levels = c("books", "play", "food", 
                                                "routines", "conv", "ac", 
                                                "gemods")), 
         id = factor(id), 
         language = factor(language), 
         speech = factor(speech))


# mlu
mlu <- read_csv("./data_demo_lena_transcripts/mlu.csv") %>% 
  mutate(activity = factor(activity, levels = c("books", "play", "food", 
                                                "routines", "conv", "ac", "ods")), 
         id = factor(id), 
         language = factor(language), 
         speech = factor(speech)) %>% 
  filter(speech == "all")


# chip
# this includes only caregivers, therefore there is no speaker column
# we exclude periods of ODS because this is about responsiveness to the child during periods of tCDS
chip <- read_csv("./data_demo_lena_transcripts/chip.csv") %>% 
  filter(activity != "ods") %>% 
  mutate(activity = factor(activity, levels = c("books", "play", "food", 
                                                "routines", "conv", "ac")), 
         id = factor(id), 
         language = factor(language)) 

str(freq)

## tibble [3,308 × 12] (S3: tbl_df/tbl/data.frame)
##  $ id           : Factor w/ 90 levels "7292","7352",..: 47 47 47 47 50 50 52 52 52 52 ...
##  $ rectime      : num [1:3308] 11923 11923 31360 31360 21499 ...
##  $ activity     : Factor w/ 7 levels "books","play",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ speaker      : chr [1:3308] "CHI" "ADULTS" "CHI" "ADULTS" ...
##  $ tokens       : num [1:3308] 30 151 35 143 58 588 42 286 33 152 ...
##  $ types        : num [1:3308] 17 70 17 65 17 199 19 53 17 59 ...
##  $ segment_num  : num [1:3308] 12 12 15 15 2 2 11 11 5 5 ...
##  $ language     : Factor w/ 2 levels "english","spanish": 1 1 1 1 1 1 1 1 1 1 ...
##  $ speech       : Factor w/ 1 level "all": 1 1 1 1 1 1 1 1 1 1 ...
##  $ dur_min      : num [1:3308] 3.55 3.55 6.57 6.57 4.71 ...
##  $ tokens_permin: num [1:3308] 8.46 42.57 5.32 21.75 12.31 ...
##  $ types_permin : num [1:3308] 4.79 19.73 2.59 9.89 3.61 ...

str(mlu)

## tibble [3,002 × 9] (S3: tbl_df/tbl/data.frame)
##  $ id         : Factor w/ 90 levels "7292","7352",..: 46 46 46 46 46 46 46 46 46 46 ...
##  $ activity   : Factor w/ 7 levels "books","play",..: 6 6 5 5 7 7 2 2 6 6 ...
##  $ speaker    : chr [1:3002] "ADULTS" "CHI" "ADULTS" "CHI" ...
##  $ segment_num: num [1:3002] 2 2 2 2 2 2 2 2 3 3 ...
##  $ words_sum  : num [1:3002] 210 66 175 43 11 16 189 47 261 78 ...
##  $ num_utt_sum: num [1:3002] 66 35 64 24 2 12 64 28 87 43 ...
##  $ mlu_w      : num [1:3002] 3.18 1.89 2.73 1.79 5.5 ...
##  $ language   : Factor w/ 2 levels "english","spanish": 1 1 1 1 1 1 1 1 1 1 ...
##  $ speech     : Factor w/ 2 levels "all","spont": 1 1 1 1 1 1 1 1 1 1 ...

str(chip)

## tibble [1,118 × 11] (S3: tbl_df/tbl/data.frame)
##  $ activity                         : Factor w/ 6 levels "books","play",..: 6 5 2 6 5 4 6 5 4 2 ...
##  $ id                               : Factor w/ 90 levels "7292","7352",..: 46 46 46 46 46 46 46 46 46 46 ...
##  $ rectime                          : num [1:1118] 15242 15242 15242 14342 14342 ...
##  $ total_adult_utt                  : num [1:1118] 68 64 65 91 43 13 50 8 65 127 ...
##  $ total_child_utt                  : num [1:1118] 46 34 33 54 17 3 14 1 29 49 ...
##  $ total_adult_resp                 : num [1:1118] 62 51 54 77 24 9 30 4 56 106 ...
##  $ total_adult_imitexp              : num [1:1118] 18 13 15 25 5 2 9 0 16 21 ...
##  $ prop_adultresp_outof_childutt    : num [1:1118] 1.35 1.5 1.64 1.43 1.41 ...
##  $ prop_adult_imitexp_outof_childutt: num [1:1118] 0.391 0.382 0.455 0.463 0.294 ...
##  $ language                         : Factor w/ 2 levels "english","spanish": 1 1 1 1 1 1 1 1 1 1 ...
##  $ segment_num                      : num [1:1118] 2 2 2 3 3 3 4 4 4 5 ...

Create dfs for ADULTS

# FREQ
freq_adult <- freq %>% 
  filter(speaker == "ADULTS")

# MLU
mlu_adult <- mlu %>% 
  filter(speaker == "ADULTS")

FREQ - Boxplots and descriptives for ADULTS

TOKENS (raw)

Freq (tokens, types)

excluded non-target children (NTC)
excluded ODS during tCDS periods
averaged across all adult speakers within activities
if an activity was not present, we did not include it (i.e., we did not impute any values to create complete cases)
0 values are included when the individual does not speak though we know they are present in the activity

# relabel for plots
freq_adult <- freq_adult %>% 
  mutate(activity = recode(activity, "conv" = "unst. conv.", "ac" = "adult-centered",
                                     "gemods" = "non_tcds"), 
         language = recode(language, "English" = "english", "Spanish" = "spanish"))


# plot for all speech
ggplot(freq_adult, aes(activity, tokens, fill = activity)) + 
  geom_boxplot() + 
  geom_jitter(alpha = .2) + 
  scale_fill_manual(values=c("darkviolet", "firebrick1", "green2", "dodgerblue1", "darkgoldenrod1", "darkgrey", "black")) + 
  facet_wrap(~ language, ncol = 1) + 
  theme(legend.position= "none") +
  theme(text = element_text(size = 35)) +
  theme(axis.text.x = element_text(angle = 20, hjust = .7)) + 
  labs(x = "", y = "")

# summarize data per participant and activity (all speech only)
freq_adult_act <- freq_adult %>% 
  group_by(id, activity) %>% 
  mutate(tokens_act = mean(tokens), 
         types_act = mean(types)) %>% 
  distinct(id, language, activity, tokens_act, types_act)

freq_adult_act_en <- freq_adult_act %>% filter(language == "english")
freq_adult_act_sp <- freq_adult_act %>% filter(language == "spanish")



# descriptives
describeBy(freq_adult_act_en$tokens_act, freq_adult_act_en$activity, mat = T, fast = T)

##     item         group1 vars  n      mean        sd    median   min       max    range      skew
## X11    1          books    1 22 480.90076 272.79828 414.25000 147.0 1115.7500 968.7500 0.6058672
## X12    2           play    1 39 234.99060 132.96267 208.33333  42.5  714.0000 671.5000 1.0915721
## X13    3           food    1 31 145.78495 124.42440  89.66667  12.0  432.0000 420.0000 0.8051817
## X14    4       routines    1 32 135.71823 107.06498 106.25000   7.0  494.0000 487.0000 1.3675315
## X15    5    unst. conv.    1 43 152.50504 100.82982 114.50000  11.0  382.7500 371.7500 0.5656557
## X16    6 adult-centered    1 45  90.56519  60.38539  87.50000   4.0  295.0000 291.0000 0.9865811
## X17    7       non_tcds    1 45 154.69407  96.87180 152.00000  15.0  418.6667 403.6667 0.7289589
##       kurtosis        se
## X11 -0.7675539 58.160789
## X12  2.0768412 21.291067
## X13 -0.7341471 22.347282
## X14  2.1046767 18.926592
## X15 -0.7104771 15.376403
## X16  1.3754902  9.001722
## X17  0.1933267 14.440795

freq_adult_en_minmax_tokens_raw <- freq_adult %>% 
  filter(language == "english") %>% 
  group_by(activity, language) %>% 
  mutate(min = min(tokens, na.rm = T), 
         max = max(tokens, na.rm = T)) %>% 
  distinct(activity, min, max)

freq_adult_en_minmax_tokens_raw

## # A tibble: 7 × 4
## # Groups:   activity, language [7]
##   language activity         min   max
##   <fct>    <fct>          <dbl> <dbl>
## 1 english  books             65  1474
## 2 english  play               0   743
## 3 english  unst. conv.        4   817
## 4 english  routines           7   692
## 5 english  food               1   630
## 6 english  adult-centered     0   397
## 7 english  non_tcds           0   945

describeBy(freq_adult_act_sp$tokens_act, freq_adult_act_sp$activity, mat = T, fast = T)

##     item         group1 vars  n     mean        sd   median       min      max    range       skew
## X11    1          books    1 20 436.9875 212.49434 425.9167 77.000000 768.0000 691.0000 -0.1097107
## X12    2           play    1 37 180.9775 136.02617 123.3333 25.000000 571.0000 546.0000  0.9509845
## X13    3           food    1 31 155.3602 119.25669 128.0000  6.000000 445.0000 439.0000  0.7248658
## X14    4       routines    1 35 166.9405 133.27275 135.0000  4.000000 635.5000 631.5000  1.4170829
## X15    5    unst. conv.    1 43 139.6674 123.38751 106.6000 13.000000 754.6667 741.6667  2.9611793
## X16    6 adult-centered    1 45 106.6911  89.50469  71.4000 21.500000 426.5000 405.0000  1.7694977
## X17    7       non_tcds    1 45 113.3085  82.01389 103.5000  4.166667 323.5000 319.3333  0.9357912
##       kurtosis       se
## X11 -1.0449431 47.51518
## X12  0.1647279 22.36256
## X13 -0.5873931 21.41913
## X14  2.3099708 22.52721
## X15 11.7931488 18.81642
## X16  2.6909840 13.34257
## X17  0.1726090 12.22591

freq_adult_sp_minmax_tokens_raw <- freq_adult %>% 
  filter(language == "spanish") %>% 
  group_by(activity, language) %>% 
  mutate(min = min(tokens, na.rm = T), 
         max = max(tokens, na.rm = T)) %>% 
  distinct(activity, min, max)

freq_adult_sp_minmax_tokens_raw

## # A tibble: 7 × 4
## # Groups:   activity, language [7]
##   language activity         min   max
##   <fct>    <fct>          <dbl> <dbl>
## 1 spanish  books             77   927
## 2 spanish  play              15   802
## 3 spanish  unst. conv.        5  1012
## 4 spanish  routines           4   937
## 5 spanish  food               6   566
## 6 spanish  adult-centered     0   566
## 7 spanish  non_tcds           0   795

TOKENS (rate per min)

freq_adult <- freq_adult %>% 
  mutate(language = recode(language, "english" = "English", "spanish" = "Spanish"))


# plot for all speech
ggplot(freq_adult, aes(activity, tokens_permin, fill = activity)) + 
  theme_classic() +
  geom_boxplot() + 
  geom_jitter(alpha = .3) + 
  scale_fill_manual(values=c("darkviolet", "firebrick1", "green2", "dodgerblue1", "darkgoldenrod1", "darkgrey", "black")) + 
  facet_wrap(~ language) + 
  theme(legend.position= "none") +
  theme(text = element_text(size = 35)) +
  theme(axis.text.x = element_text(angle = 20, hjust = .7)) + 
  labs(x = "", y = "Tokens Rate") + 
  theme(panel.spacing = unit(4, "lines"))

ggsave("./figures/boxplot_tokens_rate2.pdf", dpi = 300, width = 18, height = 8, units = "in")


# summarize data per participant and activity (all speech only)
freq_adult_act_permin <- freq_adult %>% 
  group_by(id, activity) %>% 
  mutate(tokens_permin_act = mean(tokens_permin), 
         types_permin_act = mean(types_permin)) %>% 
  distinct(id, language, activity, tokens_permin_act, types_permin_act)

freq_adult_act_permin_en <- freq_adult_act_permin %>% filter(language == "English")
freq_adult_act_permin_sp <- freq_adult_act_permin %>% filter(language == "Spanish")

# descriptives
describeBy(freq_adult_act_permin_en$tokens_permin_act, freq_adult_act_permin_en$activity, mat = T, fast = T)

##     item         group1 vars  n     mean       sd   median       min       max     range       skew
## X11    1          books    1 22 97.05612 25.83716 99.10089 32.157566 139.35186 107.19429 -0.4652193
## X12    2           play    1 39 63.00664 24.36993 58.04526 19.809892 122.54285 102.73296  0.6362527
## X13    3           food    1 31 63.20593 33.34061 56.34873 17.818695 191.38756 173.56887  1.7407592
## X14    4       routines    1 32 70.52775 27.24599 66.26826 28.267219 157.18563 128.91841  1.1449756
## X15    5    unst. conv.    1 43 76.14884 35.36817 73.18092 10.266822 224.58716 214.32033  1.4451765
## X16    6 adult-centered    1 45 85.01633 73.45787 77.63141 19.605147 534.52116 514.91601  5.0378276
## X17    7       non_tcds    1 45 36.98633 21.17975 39.00190  4.063964  87.65849  83.59453  0.5225623
##        kurtosis        se
## X11 -0.09411375  5.508500
## X12  0.06339446  3.902312
## X13  4.60071546  5.988150
## X14  1.57756298  4.816457
## X15  5.13425782  5.393596
## X16 28.20995719 10.950453
## X17 -0.45795070  3.157290

freq_adult_en_minmax_tokens_rate <- freq_adult %>% 
  filter(language == "English") %>% 
  group_by(activity, language) %>% 
  mutate(min = min(tokens_permin, na.rm = T), 
         max = max(tokens_permin, na.rm = T)) %>% 
  distinct(activity, min, max)

freq_adult_en_minmax_tokens_rate

## # A tibble: 7 × 4
## # Groups:   activity, language [7]
##   language activity         min   max
##   <fct>    <fct>          <dbl> <dbl>
## 1 English  books          21.7   149.
## 2 English  play            0     183.
## 3 English  unst. conv.     2.93  225.
## 4 English  routines       21.4   157.
## 5 English  food           12.3   191.
## 6 English  adult-centered  0     535.
## 7 English  non_tcds        0     167.

describeBy(freq_adult_act_permin_sp$tokens_permin_act, freq_adult_act_permin_sp$activity, mat = T, fast = T)

##     item         group1 vars  n     mean       sd   median       min       max     range       skew
## X11    1          books    1 20 69.70156 22.42204 71.71504 33.415438 114.94578  81.53034 0.11976814
## X12    2           play    1 37 51.27875 20.41805 52.27932  8.535673 118.63334 110.09767 0.62021093
## X13    3           food    1 31 37.39510 15.81841 37.57642  9.649060  71.37418  61.72512 0.07164798
## X14    4       routines    1 35 59.05829 22.87843 58.32433 13.861479 118.88422 105.02275 0.42422784
## X15    5    unst. conv.    1 43 59.56066 23.87057 57.76884 24.971949 137.97635 113.00440 0.98462727
## X16    6 adult-centered    1 45 56.17319 33.33514 53.78916 14.119577 161.15476 147.03518 1.17089383
## X17    7       non_tcds    1 45 31.05852 17.21041 28.27339  2.157086  71.16063  69.00355 0.29725236
##       kurtosis       se
## X11 -0.8333837 5.013720
## X12  1.4100577 3.356707
## X13 -0.7574487 2.841070
## X14  0.2797878 3.867161
## X15  1.0899492 3.640229
## X16  1.4051846 4.969310
## X17 -0.8203294 2.565577

freq_adult_sp_minmax_tokens_rate <- freq_adult %>% 
  filter(language == "Spanish") %>% 
  group_by(activity, language) %>% 
  mutate(min = min(tokens_permin, na.rm = T), 
         max = max(tokens_permin, na.rm = T)) %>% 
  distinct(activity, min, max)

freq_adult_sp_minmax_tokens_rate

## # A tibble: 7 × 4
## # Groups:   activity, language [7]
##   language activity         min   max
##   <fct>    <fct>          <dbl> <dbl>
## 1 Spanish  books          33.4  115. 
## 2 Spanish  play            8.54 128. 
## 3 Spanish  unst. conv.    10.4  140. 
## 4 Spanish  routines       13.3  183. 
## 5 Spanish  food            9.65  91.1
## 6 Spanish  adult-centered  0    343. 
## 7 Spanish  non_tcds        0    153.

TYPES (raw)

# df for just all speech
ggplot(freq_adult, aes(activity, types, fill = activity)) +
  geom_boxplot() + 
  geom_jitter(alpha = .2) + 
  scale_fill_manual(values=c("darkviolet", "firebrick1", "green2", "dodgerblue1", "darkgoldenrod1", "darkgrey", "black")) + 
  facet_wrap(~ language) + 
  theme(legend.position= "none") +
  theme(text = element_text(size = 35)) +
  theme(axis.text.x = element_text(angle = 20, hjust = .7)) + 
  labs(x = "", y = "")

# descriptives
describeBy(freq_adult_act_en$types_act, freq_adult_act_en$activity, mat = T, fast = T)

##     item         group1 vars  n      mean       sd    median  min      max    range      skew
## X11    1          books    1 22 143.90227 66.70063 116.50000 56.0 277.7500 221.7500 0.4732497
## X12    2           play    1 39  85.90256 36.92368  81.50000 27.5 197.0000 169.5000 0.8994632
## X13    3           food    1 31  64.56989 42.01631  50.00000 12.0 177.0000 165.0000 0.6751602
## X14    4       routines    1 32  58.62083 31.98695  52.00000  7.0 144.0000 137.0000 0.6238625
## X15    5    unst. conv.    1 43  65.25039 36.72048  53.33333 10.0 163.0000 153.0000 0.6798436
## X16    6 adult-centered    1 45  44.60630 24.26759  43.40000  4.0 129.0000 125.0000 0.8473480
## X17    7       non_tcds    1 45  79.91481 44.48101  71.33333 10.5 199.8333 189.3333 0.4831218
##        kurtosis        se
## X11 -1.12778313 14.220623
## X12  1.00289109  5.912521
## X13 -0.46931520  7.546351
## X14  0.06720687  5.654547
## X15 -0.18936585  5.599821
## X16  1.60694389  3.617598
## X17 -0.42847167  6.630837

freq_adult_en_minmax_types_raw <- freq_adult %>% 
  filter(language == "English") %>% 
  group_by(activity, language) %>% 
  mutate(min = min(types, na.rm = T), 
         max = max(types, na.rm = T)) %>% 
  distinct(activity, min, max)

freq_adult_en_minmax_types_raw

## # A tibble: 7 × 4
## # Groups:   activity, language [7]
##   language activity         min   max
##   <fct>    <fct>          <dbl> <dbl>
## 1 English  books             38   369
## 2 English  play               0   253
## 3 English  unst. conv.        4   267
## 4 English  routines           7   246
## 5 English  food               1   181
## 6 English  adult-centered     0   159
## 7 English  non_tcds           0   351

describeBy(freq_adult_act_sp$types_act, freq_adult_act_sp$activity, mat = T, fast = T)

##     item         group1 vars  n      mean       sd    median  min      max    range      skew
## X11    1          books    1 20 144.82500 64.74266 138.50000 44.0 268.5000 224.5000 0.2285406
## X12    2           play    1 37  62.33784 35.84100  54.66667 14.0 157.0000 143.0000 0.6215909
## X13    3           food    1 31  65.50538 36.39642  69.00000  5.0 133.0000 128.0000 0.1241942
## X14    4       routines    1 35  65.69762 33.60470  65.00000  4.0 141.5000 137.5000 0.3473994
## X15    5    unst. conv.    1 43  59.88527 32.98391  52.00000  8.0 160.6667 152.6667 0.8136958
## X16    6 adult-centered    1 45  48.42926 26.92731  40.60000 15.0 121.0000 106.0000 1.0463985
## X17    7       non_tcds    1 45  59.29000 36.12099  55.16667  3.5 137.8333 134.3333 0.5162437
##        kurtosis        se
## X11 -1.03063121 14.476900
## X12 -0.30729538  5.892225
## X13 -1.22818040  6.536989
## X14 -0.46882830  5.680232
## X15  0.38015627  5.030000
## X16  0.08898838  4.014086
## X17 -0.42162311  5.384600

freq_adult_sp_minmax_types_raw <- freq_adult %>% 
  filter(language == "Spanish") %>% 
  group_by(activity, language) %>% 
  mutate(min = min(types, na.rm = T), 
         max = max(types, na.rm = T)) %>% 
  distinct(activity, min, max)

freq_adult_sp_minmax_types_raw

## # A tibble: 7 × 4
## # Groups:   activity, language [7]
##   language activity         min   max
##   <fct>    <fct>          <dbl> <dbl>
## 1 Spanish  books             44   361
## 2 Spanish  play               7   186
## 3 Spanish  unst. conv.        5   219
## 4 Spanish  routines           4   175
## 5 Spanish  food               5   166
## 6 Spanish  adult-centered     0   212
## 7 Spanish  non_tcds           0   289

TYPES (rate per min)

# plot for all speech
ggplot(freq_adult, aes(activity, types_permin, fill = activity)) + 
  theme_classic() +
  geom_boxplot() + 
  geom_jitter(alpha = .3) + 
  scale_fill_manual(values=c("darkviolet", "firebrick1", "green2", "dodgerblue1", "darkgoldenrod1", "darkgrey", "black")) + 
  facet_wrap(~ language) + 
  theme(legend.position= "none") +
  theme(text = element_text(size = 35)) +
  theme(axis.text.x = element_text(angle = 20, hjust = .7)) + 
  labs(x = "", y = "Types Rate")

ggsave("./figures/boxplot_types_rate2.pdf", dpi = 300, width = 18, height = 8, units = "in")


# descriptives
describeBy(freq_adult_act_permin_en$types_permin_act, freq_adult_act_permin_en$activity, mat = T, fast = T)

##     item         group1 vars  n     mean       sd   median       min       max     range      skew
## X11    1          books    1 22 32.06383  9.75467 32.15751 14.809274  48.75449  33.94522 0.1938847
## X12    2           play    1 39 27.07891 12.32556 24.04612 12.245120  66.31838  54.07326 1.5066690
## X13    3           food    1 31 39.82161 34.99837 30.28977  9.280570 191.38756 182.10699 2.7880293
## X14    4       routines    1 32 41.75257 25.57367 35.05380 14.409224 157.18563 142.77640 2.8301947
## X15    5    unst. conv.    1 43 41.83628 27.40421 36.04217  5.648173 176.14679 170.49862 2.8015041
## X16    6 adult-centered    1 45 61.14224 75.86224 46.13497 10.605640 534.52116 523.91552 5.3822148
## X17    7       non_tcds    1 45 20.66782 11.41460 20.40530  2.392211  48.90393  46.51172 0.5745082
##       kurtosis        se
## X11 -1.0590251  2.079703
## X12  2.2094453  1.973670
## X13  8.8782125  6.285893
## X14 10.1753362  4.520830
## X15 10.9712805  4.179103
## X16 30.7217214 11.308874
## X17 -0.3456499  1.701588

freq_adult_en_minmax_types_rate <- freq_adult %>% 
  filter(language == "English") %>% 
  group_by(activity, language) %>% 
  mutate(min = min(types_permin, na.rm = T), 
         max = max(types_permin, na.rm = T)) %>% 
  distinct(activity, min, max)

freq_adult_en_minmax_types_rate

## # A tibble: 7 × 4
## # Groups:   activity, language [7]
##   language activity         min   max
##   <fct>    <fct>          <dbl> <dbl>
## 1 English  books           9.89  64.4
## 2 English  play            0    149. 
## 3 English  unst. conv.     2.93 176. 
## 4 English  routines       11.2  157. 
## 5 English  food            8.50 191. 
## 6 English  adult-centered  0    535. 
## 7 English  non_tcds        0     92.2

describeBy(freq_adult_act_permin_sp$types_permin_act, freq_adult_act_permin_sp$activity, mat = T, fast = T)

##     item         group1 vars  n     mean        sd   median       min       max     range      skew
## X11    1          books    1 20 25.19438  9.378178 25.62804 10.866809  49.60044  38.73363 0.6192718
## X12    2           play    1 37 21.98845 11.038304 19.62847  4.405509  66.43467  62.02916 1.7173151
## X13    3           food    1 31 19.09226  8.249124 17.82146  8.576942  39.99831  31.42137 0.8860032
## X14    4       routines    1 35 30.57141 14.693167 28.80377  9.082726  73.61036  64.52764 1.1037506
## X15    5    unst. conv.    1 43 32.41628 16.819376 26.79962 10.061461 100.19711  90.13565 1.7190575
## X16    6 adult-centered    1 45 38.00249 28.213513 33.91582  7.901010 137.53863 129.63762 1.8238166
## X17    7       non_tcds    1 45 18.23642  9.936926 16.33417  1.666453  43.46532  41.79886 0.5847300
##        kurtosis       se
## X11  0.07872493 2.097024
## X12  4.88857397 1.814686
## X13 -0.11762887 1.481586
## X14  1.11375487 2.483598
## X15  4.05465209 2.564931
## X16  3.43928396 4.205822
## X17 -0.16215447 1.481309

freq_adult_sp_minmax_types_rate <- freq_adult %>% 
  filter(language == "Spanish") %>% 
  group_by(activity, language) %>% 
  mutate(min = min(types_permin, na.rm = T), 
         max = max(types_permin, na.rm = T)) %>% 
  distinct(activity, min, max)

freq_adult_sp_minmax_types_rate

## # A tibble: 7 × 4
## # Groups:   activity, language [7]
##   language activity         min   max
##   <fct>    <fct>          <dbl> <dbl>
## 1 Spanish  books          10.9   49.6
## 2 Spanish  play            4.41  71.2
## 3 Spanish  unst. conv.     6.68 100. 
## 4 Spanish  routines        8.17 171. 
## 5 Spanish  food            5.38  55.7
## 6 Spanish  adult-centered  0    343. 
## 7 Spanish  non_tcds        0    130.

MLU

excluded NTC
KEPT ODS during tCDS periods
averaged across all adult speakers
if an activity was not present, we did not include it (i.e., we did not impute any values to create complete cases)
we excluded observations when there were 0 number of utterances, since we cannot calculate an MLU when there are no utterances

# relabel for plots
mlu_adult <- mlu_adult %>% 
  mutate(activity = recode(activity, "conv" = "unst. conv.", "ac" = "adult-centered",
                           "ods" = "non_tcds"), 
         language = recode(language, "english" = "English", "spanish" = "Spanish"))


# plot
ggplot(mlu_adult, aes(activity, mlu_w, fill = activity)) + 
  theme_classic() +
  geom_boxplot() + 
  geom_jitter(alpha = .3) + 
  scale_fill_manual(values=c("darkviolet", "firebrick1", "green2", "dodgerblue1", "darkgoldenrod1", "darkgrey", "black")) + 
  facet_wrap(~ language) + 
  theme(legend.position= "none") +
  theme(text = element_text(size = 35)) +
  theme(axis.text.x = element_text(angle = 20, hjust = .7)) + 
  labs(x = "", y = "MLUw")

ggsave("./figures/boxplot_mluw2.pdf", dpi = 300, width = 18, height = 8, units = "in")


# summarize data per participant and activity (all speech only)
mlu_adult_act <- mlu_adult %>% 
  group_by(id, activity) %>% 
  mutate(mlu_w_act = mean(mlu_w)) %>% 
  distinct(id, language, activity, mlu_w_act)

mlu_adult_act_en <- mlu_adult_act %>% filter(language == "English")
mlu_adult_act_sp <- mlu_adult_act %>% filter(language == "Spanish")

# descriptives
describeBy(mlu_adult_act_en$mlu_w_act, mlu_adult_act_en$activity, mat = T, fast = T)

##     item         group1 vars  n     mean        sd   median      min      max    range        skew
## X11    1          books    1 22 4.860163 1.2386474 4.849059 2.843305 7.287671 4.444366  0.38589952
## X12    2           play    1 39 3.544910 0.7799806 3.361351 2.117647 5.428571 3.310924  0.37501840
## X13    3           food    1 31 3.508170 0.8406000 3.268261 2.125000 5.328869 3.203869  0.35194344
## X14    4       routines    1 32 3.665860 0.7357554 3.610858 2.200000 5.670455 3.470455  0.54603786
## X15    5    unst. conv.    1 43 3.763843 0.8423445 3.772650 1.500000 5.769231 4.269231 -0.32572952
## X16    6 adult-centered    1 45 3.524902 0.7379635 3.407890 2.248271 5.668605 3.420334  0.87918296
## X17    7       non_tcds    1 45 4.150646 0.8011976 4.165139 2.542017 5.770249 3.228232  0.04840642
##        kurtosis        se
## X11 -0.76500796 0.2640805
## X12 -0.52053041 0.1248969
## X13 -0.75764546 0.1509762
## X14  0.06066605 0.1300644
## X15  0.41938715 0.1284563
## X16  0.83025135 0.1100091
## X17 -0.82614709 0.1194355

mlu_adult_en_minmax <- mlu_adult %>% 
  filter(language == "English") %>% 
  group_by(activity, language) %>% 
  mutate(min = min(mlu_w, na.rm = T), 
         max = max(mlu_w, na.rm = T)) %>% 
  distinct(activity, min, max)

mlu_adult_en_minmax

## # A tibble: 7 × 4
## # Groups:   activity, language [7]
##   language activity         min   max
##   <fct>    <fct>          <dbl> <dbl>
## 1 English  adult-centered  1     9   
## 2 English  unst. conv.     1     6.67
## 3 English  non_tcds        1     8   
## 4 English  play            1.33  5.69
## 5 English  routines        2.2   8   
## 6 English  books           1.51  8.73
## 7 English  food            1     6.06

describeBy(mlu_adult_act_sp$mlu_w_act, mlu_adult_act_sp$activity, mat = T, fast = T)

##     item         group1 vars  n     mean        sd   median      min      max    range       skew
## X11    1          books    1 20 3.822411 1.3037314 3.758225 1.864407 6.566855 4.702448  0.4630879
## X12    2           play    1 37 2.698639 0.7463847 2.542857 1.422360 4.545918 3.123558  0.4675977
## X13    3           food    1 31 2.772200 0.6975939 2.746032 1.391304 4.230769 2.839465 -0.0686621
## X14    4       routines    1 35 3.029701 0.7381221 3.000000 1.791667 4.485947 2.694281  0.2233092
## X15    5    unst. conv.    1 43 3.054863 0.5741442 2.821722 2.004357 4.589286 2.584928  0.7193086
## X16    6 adult-centered    1 45 2.756766 0.6920525 2.682044 1.666667 5.019476 3.352810  0.8007356
## X17    7       non_tcds    1 45 3.499380 0.7220236 3.461212 1.890000 5.202889 3.312889  0.1197576
##       kurtosis         se
## X11 -0.7168780 0.29152319
## X12 -0.4561800 0.12270488
## X13 -0.5377128 0.12529156
## X14 -1.0421283 0.12476540
## X15 -0.2066920 0.08755616
## X16  0.6194120 0.10316510
## X17 -0.2901765 0.10763292

mlu_adult_sp_minmax <- mlu_adult %>% 
  filter(language == "Spanish") %>% 
  group_by(activity, language) %>% 
  mutate(min = min(mlu_w, na.rm = T), 
         max = max(mlu_w, na.rm = T)) %>% 
  distinct(activity, min, max)

mlu_adult_sp_minmax

## # A tibble: 7 × 4
## # Groups:   activity, language [7]
##   language activity         min   max
##   <fct>    <fct>          <dbl> <dbl>
## 1 Spanish  books           1.86  8.36
## 2 Spanish  play            1.21  6.48
## 3 Spanish  unst. conv.     1     5.33
## 4 Spanish  routines        1.58  5   
## 5 Spanish  food            1.39  4.32
## 6 Spanish  adult-centered  1     6.22
## 7 Spanish  non_tcds        1     7.9

CHIP

Responses (RAW and PROP)

CHIP (responses, imitations/expansions; these are utterances that follow a child’s utterance, within a 5 utterance window)

excluded NTC
excluded overlapping ODS during CDS periods
averaged across all adult speakers
if an activity was not present, we did not include it (i.e., we did not impute any values to create complete cases)
we keep 0 values, which are because the adults did not respond even though the child did have utterances [e.g., numerator is 0 but denominator is a value]; if a child did not have utterances, then adult responses could not be calculated so the number of observations differ here from freq and mlu
greater than 1 = caregiver had more utterances in response to the child; less than 1 = child had more utterances than caregiver responses

# create dfs
chip2 <- chip %>% 
  mutate(activity = recode(activity, "conv" = "unst. conv.", "ac" = "adult-centered"), 
         language = recode(language, "english" = "English", "spanish" = "Spanish"))


# plot - total adult responses
ggplot(chip2, aes(activity, total_adult_resp, fill = activity)) + 
  geom_boxplot() + 
  geom_jitter(alpha = .2) + 
  scale_fill_manual(values=c("darkviolet", "firebrick1", "green2", "dodgerblue1", "darkgoldenrod1", "darkgrey")) + 
  facet_wrap(~ language, ncol = 1) + 
  theme(legend.position= "none") +
  theme(text = element_text(size = 35)) +
  theme(axis.text.x = element_text(angle = 20, hjust = .7)) + 
  labs(x = "", y = "")

# plot - prop of adult resp/child utt
ggplot(chip2, aes(activity, prop_adultresp_outof_childutt, fill = activity)) + 
  theme_classic() +
  geom_boxplot() + 
  geom_jitter(alpha = .3) + 
  scale_fill_manual(values=c("darkviolet", "firebrick1", "green2", "dodgerblue1", "darkgoldenrod1", "darkgrey")) + 
  facet_wrap(~ language) + 
  theme(legend.position= "none") +
  theme(text = element_text(size = 35)) +
  geom_hline(yintercept = 1) + # caregiver and child = equal utts
  theme(axis.text.x = element_text(angle = 20, hjust = .7)) + 
  labs(x = "", y = "Prop Responses")

ggsave("./figures/boxplot_prop_resp2.pdf", dpi = 300, width = 18, height = 8, units = "in")


# summarize data per participant and activity (all speech only)
chip_act <- chip %>% 
  group_by(id, activity) %>% 
  mutate(prop_resp_act = mean(prop_adultresp_outof_childutt), 
         prop_imitexp_act = mean(prop_adult_imitexp_outof_childutt)) %>% 
  distinct(id, language, activity, prop_resp_act, prop_imitexp_act)

chip_act_en <- chip_act %>% filter(language == "english")
chip_act_sp <- chip_act %>% filter(language == "spanish")


# descriptives
describeBy(chip_act_en$prop_resp_act, chip_act_en$activity, mat = T, fast = T)

##     item   group1 vars  n     mean        sd   median       min      max    range        skew
## X11    1    books    1 22 2.378070 0.7766933 2.342175 1.1864443 4.366667 3.180222  0.69916859
## X12    2     play    1 36 2.074418 0.6357208 1.925870 1.1301724 4.666667 3.536494  1.95571413
## X13    3     food    1 29 1.914779 0.5461100 1.944444 0.8141026 3.066667 2.252564  0.14337467
## X14    4 routines    1 28 1.856904 0.5897762 1.777778 0.8000000 3.333333 2.533333  0.57225823
## X15    5     conv    1 36 1.866984 0.5430045 1.849124 0.6250000 2.933333 2.308333 -0.03709437
## X16    6       ac    1 22 1.855095 0.5401576 1.815131 0.9909502 3.248677 2.257727  0.54471173
##        kurtosis         se
## X11  0.03584900 0.16559157
## X12  5.35965871 0.10595346
## X13 -0.49085257 0.10141009
## X14 -0.07450000 0.11145723
## X15 -0.55933392 0.09050074
## X16 -0.07700799 0.11516198

chip_adult_en_minmax_propresp <- chip %>% 
  filter(language == "english") %>% 
  group_by(activity, language) %>% 
  mutate(min = min(prop_adultresp_outof_childutt, na.rm = T), 
         max = max(prop_adultresp_outof_childutt, na.rm = T)) %>% 
  distinct(activity, min, max)

chip_adult_en_minmax_propresp

## # A tibble: 6 × 4
## # Groups:   activity, language [6]
##   language activity   min   max
##   <fct>    <fct>    <dbl> <dbl>
## 1 english  ac       0      5   
## 2 english  conv     0.571  5   
## 3 english  play     0.667  4.67
## 4 english  routines 0.5    3.8 
## 5 english  books    0.859  5   
## 6 english  food     0      5

describeBy(chip_act_sp$prop_resp_act, chip_act_sp$activity, mat = T, fast = T)

##     item   group1 vars  n     mean        sd   median       min      max    range       skew
## X11    1    books    1 20 2.125135 0.4832411 2.109579 1.0617284 3.055556 1.993827 -0.3732188
## X12    2     play    1 31 1.927563 0.6742332 1.750000 0.8125000 3.500000 2.687500  0.5481160
## X13    3     food    1 30 1.762135 0.7100370 1.620192 0.3333333 3.314815 2.981481  0.3986445
## X14    4 routines    1 31 1.904239 0.5335071 1.900000 1.0000000 3.166667 2.166667  0.3703352
## X15    5     conv    1 35 1.818201 0.5605011 1.683208 0.8823529 3.119048 2.236695  0.6188577
## X16    6       ac    1 25 1.589711 0.6019761 1.524806 0.7083333 3.750000 3.041667  1.6448000
##       kurtosis         se
## X11 -0.2278578 0.10805599
## X12 -0.5407181 0.12109586
## X13 -0.3813870 0.12963442
## X14 -0.4500460 0.09582070
## X15 -0.4151453 0.09474197
## X16  4.0945489 0.12039523

chip_adult_sp_minmax_propresp <- chip %>% 
  filter(language == "spanish") %>% 
  group_by(activity, language) %>% 
  mutate(min = min(prop_adultresp_outof_childutt, na.rm = T), 
         max = max(prop_adultresp_outof_childutt, na.rm = T)) %>% 
  distinct(activity, min, max)

chip_adult_sp_minmax_propresp

## # A tibble: 6 × 4
## # Groups:   activity, language [6]
##   language activity   min   max
##   <fct>    <fct>    <dbl> <dbl>
## 1 spanish  ac       0      5   
## 2 spanish  conv     0.7    5   
## 3 spanish  play     0.765  5   
## 4 spanish  food     0.333  4   
## 5 spanish  routines 1      5   
## 6 spanish  books    1.06   3.06

Imitations/Expansions (RAW and PROP)

# plot
ggplot(chip2, aes(activity, total_adult_imitexp, fill = activity)) + 
  geom_boxplot() + 
  geom_jitter(alpha = .2) + 
  scale_fill_manual(values=c("darkviolet", "firebrick1", "green2", "dodgerblue1", "darkgoldenrod1", "darkgrey")) + 
  facet_wrap(~ language, ncol = 1) + 
  theme(legend.position= "none") +
  theme(text = element_text(size = 35)) +
  theme(axis.text.x = element_text(angle = 20, hjust = .7)) + 
  labs(x = "", y = "")

# plot
ggplot(chip2, aes(activity, prop_adult_imitexp_outof_childutt, fill = activity)) + 
  theme_classic() + 
  geom_boxplot() + 
  geom_jitter(alpha = .3) + 
  scale_fill_manual(values=c("darkviolet", "firebrick1", "green2", "dodgerblue1", "darkgoldenrod1", "darkgrey")) + 
  facet_wrap(~ language) + 
  theme(legend.position= "none") +
  theme(text = element_text(size = 35)) +
  geom_hline(yintercept = 1) + # caregiver and child = equal utts
  theme(axis.text.x = element_text(angle = 20, hjust = .7)) + 
  labs(x = "", y = "Prop Imitations/Expansions")

ggsave("./figures/boxplot_prop_imitexp2.pdf", dpi = 300, width = 18, height = 8, units = "in")


# descriptives
describeBy(chip_act_en$prop_imitexp_act, chip_act_en$activity, mat = T, fast = T)

##     item   group1 vars  n      mean        sd    median        min       max     range        skew
## X11    1    books    1 22 0.4822402 0.2032150 0.4606055 0.07142857 0.8000000 0.7285714 -0.17741053
## X12    2     play    1 36 0.4082746 0.2249434 0.4011154 0.00000000 1.1157407 1.1157407  0.96845872
## X13    3     food    1 29 0.3716799 0.1958653 0.3809524 0.00000000 1.0000000 1.0000000  0.91498868
## X14    4 routines    1 28 0.3527558 0.1764344 0.3737981 0.00000000 0.6746032 0.6746032 -0.08511947
## X15    5     conv    1 36 0.3490170 0.1925299 0.3014535 0.00000000 0.8888889 0.8888889  0.68222826
## X16    6       ac    1 22 0.3648964 0.1674221 0.3302305 0.18972991 0.8666667 0.6769368  1.33927429
##       kurtosis         se
## X11 -0.8102586 0.04332558
## X12  1.4668965 0.03749056
## X13  1.7868059 0.03637128
## X14 -0.7390449 0.03334296
## X15  0.1562880 0.03208831
## X16  1.3695376 0.03569452

chip_adult_en_minmax_propimitexp <- chip %>% 
  filter(language == "english") %>% 
  group_by(activity, language) %>% 
  mutate(min = min(prop_adult_imitexp_outof_childutt, na.rm = T), 
         max = max(prop_adult_imitexp_outof_childutt, na.rm = T)) %>% 
  distinct(activity, min, max)

chip_adult_en_minmax_propimitexp

## # A tibble: 6 × 4
## # Groups:   activity, language [6]
##   language activity    min   max
##   <fct>    <fct>     <dbl> <dbl>
## 1 english  ac       0       4   
## 2 english  conv     0       2   
## 3 english  play     0       1.38
## 4 english  routines 0       1   
## 5 english  books    0.0714  1   
## 6 english  food     0       1

describeBy(chip_act_sp$prop_imitexp_act, chip_act_sp$activity, mat = T, fast = T)

##     item   group1 vars  n      mean        sd    median    min       max     range      skew
## X11    1    books    1 20 0.4052845 0.2414756 0.3428358 0.0625 1.0000000 0.9375000 0.7095279
## X12    2     play    1 31 0.3536352 0.2224772 0.3333333 0.0000 1.1700000 1.1700000 1.4354164
## X13    3     food    1 30 0.3333968 0.2401772 0.3088362 0.0000 1.0000000 1.0000000 0.9717588
## X14    4 routines    1 31 0.3949755 0.2920670 0.3562162 0.0000 1.0322636 1.0322636 0.5266545
## X15    5     conv    1 35 0.3856342 0.2314362 0.3640005 0.0000 0.9523810 0.9523810 0.7019027
## X16    6       ac    1 25 0.3015461 0.1634934 0.2982069 0.0000 0.7603067 0.7603067 0.8202077
##        kurtosis         se
## X11 -0.32529803 0.05399559
## X12  3.62662516 0.03995808
## X13  0.58774011 0.04385016
## X14 -0.74778343 0.05245678
## X15 -0.07136292 0.03911986
## X16  0.67007692 0.03269868

chip_adult_sp_minmax_propimitexp <- chip %>% 
  filter(language == "spanish") %>% 
  group_by(activity, language) %>% 
  mutate(min = min(prop_adult_imitexp_outof_childutt, na.rm = T), 
         max = max(prop_adult_imitexp_outof_childutt, na.rm = T)) %>% 
  distinct(activity, min, max)

chip_adult_sp_minmax_propimitexp

## # A tibble: 6 × 4
## # Groups:   activity, language [6]
##   language activity   min   max
##   <fct>    <fct>    <dbl> <dbl>
## 1 spanish  ac           0  1.67
## 2 spanish  conv         0  2   
## 3 spanish  play         0  2   
## 4 spanish  food         0  1   
## 5 spanish  routines     0  3   
## 6 spanish  books        0  1

Number of Speakers

# this was extracted using speaker roles from the FREQ output
# in a prior code, the n_speakers was calculated to reflect the number of speakers for the speaker role (child, adults), during the respective activity and segment num
# removing kwalods because these are overlapping utterances during tCDS activities

num_speakers <- read_csv("./data_demo_lena_transcripts/num_speakers.csv") %>% 
  filter(Speaker2 == "ADULTS" & activity != "kwalods") %>% 
  distinct(id, activity, segment_num, language, n_speakers) %>% 
  mutate(bin_n_adult_speakers = ifelse(n_speakers > 1, "more_than1", "1_adult")) %>% 
  group_by(activity) %>% 
  mutate(n_instances_by_activity = n()) %>% 
  ungroup() %>% 
  group_by(activity, bin_n_adult_speakers) %>% 
  mutate(n_by_bin = n())


num_speakers %>% 
  ungroup() %>% 
  distinct(activity, bin_n_adult_speakers, n_by_bin, n_instances_by_activity) %>% 
  mutate(prop = n_by_bin / n_instances_by_activity)

## # A tibble: 14 × 5
##    activity bin_n_adult_speakers n_by_bin n_instances_by_activity  prop
##    <chr>    <chr>                   <int>                   <int> <dbl>
##  1 books    1_adult                    67                      84 0.798
##  2 books    more_than1                 17                      84 0.202
##  3 play     1_adult                   133                     181 0.735
##  4 play     more_than1                 48                     181 0.265
##  5 conv     1_adult                   157                     254 0.618
##  6 conv     more_than1                 97                     254 0.382
##  7 routines 1_adult                    91                     121 0.752
##  8 routines more_than1                 30                     121 0.248
##  9 food     1_adult                    74                     115 0.643
## 10 food     more_than1                 41                     115 0.357
## 11 ac       1_adult                   215                     363 0.592
## 12 ac       more_than1                148                     363 0.408
## 13 gemods   1_adult                   259                     536 0.483
## 14 gemods   more_than1                277                     536 0.517

Descriptives of Caregiver Speech - Transcription data

Janet Bang

April 27, 2025